|
|
|
|
|
|
| Clear images and colors | Post date: 2018/11/5 | |
H-SCAN:
High-performance computing (HPC) is the use of parallel processing for running advanced application programs efficiently, reliably and quickly. In this group, different researches is doing such as gang scheduling. The aim of scheduling algorithms in distributed computing systems is mapping the tasks to the processors and making decisions to execute them in a certain way to optimize the overall performance of the system. Adding a time-sharing dimension to primitive approach was a solution to reduce the processing resource dissipations. This measure, which is called gang scheduling (GS), is useful to schedule the workloads whose jobs consist of parallel tasks which need to communicate with each other frequently. In GS, the tasks of one or more jobs are grouped together into a gang. All tasks of a gang are executed on a set of processors simultaneously so there is a one-to-one mapping between tasks and processors at any moment.
Graph-based methods have been applied in a variety of fields, such as computer science, bioinformatics, chemistry, and physics. In these fields, efficient graph processing method is necessary when dealing with million/billion vertices and edges. Multi-GPUs nodes can become platform for graph processing. Owing to its characteristics of strong parallel capability and high bandwidth, GPU is able to meet high-performance of graph processing. Therefore, many researchers have begun to pay increasing attention on large-scale graph processing on GPU in recent years. My work is social network graph processing with GPU.
PhD Students:
1-Mastoori
2-Moeni
3-Mollajafari
MSc Students:
1-Doozbakhshan
2-Bakhshayeshi
3-Doostmohammadi
PhD Alumni:
1-Mollajafari
MSc Alumni:
1-Nazarian
2-Arghand
3-Govahi
4-Amir
Publications:
Mollajafari, M. and Shahhoseini, H.S., 2016. Cost-Optimized GA-Based Heuristic for Scheduling Time-Constrained Workflow Applications in Infrastructure Clouds Using an Innovative Feasibility-Assured Decoding Mechanism. J. Inf. Sci. Eng., 32(6), pp.1541-1560.
Amir, H. and Shahhoseini, H.S., 2013. Improving CompactMatrix phase in gang scheduling by changing transference condition and utilizing exchange. The Journal of Supercomputing, 66(3), pp.1707-1728. |
|
|
| |